The goal of this project is to develop an intraperitoneally implanted, closed-loop glucose-monitoring / insulin- delivery device (aka artificial pancreas) for people with type 1 diabetes. An artificial pancreas could improve patients short- and long-term health and quality of life, and it could decrease overall healthcare costs. However, most current artificial-pancreas prototypes use subcutaneous insulin pumps and subcutaneous glucose sensors, which are associated with significant lag times (~60 minutes and ~10 minutes, respectively). Furthermore, insulin infusion sites and sensors require replacement every several days, contributing to variability in insulin absorption and glucose measurement. These lags and variabilities impose a ceiling on the performance of any artificial pancreas that uses subcutaneous insulin delivery and glucose sensing. Fortunately, these challenges could be overcome by an implantable artificial pancreas that uses both intraperitoneal insulin delivery and intraperitoneal glucose sensing - an IP-IP AP. Compared to subcutaneous insulin, intraperitoneal insulin is faster and more predictable in both onset and offset; compared to subcutaneous sensing, intraperitoneal sensing appears more accurate and timely for tracking blood glucose changes. Intraday variability is minimal in the intraperitoneal environment, and re-implantation would be required only once per year (with the pump re-filled transcutaneously once every three months). Looking toward real-world impact, the ultimate goal of this research is development of a fully closed-loop, commercializable device with total costs that are identical (or lower) compared to current pump/CGM therapy. Development of an IP-IP AP will be a multidisciplinary effort. Thus, the project team includes: a company developing fluorescence-based implantable glucose sensors; a company developing intraperitoneally implantable, miniaturized insulin pumps; a company developing a heat-stable, highly concentrated insulin analog; a company with expertise in intraperitoneal glucose sensing; an academic institution with expertise in glucose-control algorithms; and a clinical research institut with expertise in diabetes technology. The project has five specific aims: 1) To adapt an existing fluorescence-based glucose sensor for integration into an intraperitoneal insulin catheter, and to build roughly 50 sensor-catheters for animal and human studies; 2) To develop and manufacture a next-generation implantable, MEMS-based insulin pump that can be wirelessly operated as part of a closed-loop system; 3) To confirm that the heat-stable insulin analog is compatible with use in an intraperitoneal pump at 37 C for up to one year, and to characterize the analog's intraperitoneal PK and PD; 4) To optimize a closed-loop control algorithm for intraperitoneal sensing and intraperitoneal insulin delivery, and to validate this algorithm with i silico simulations; 5) To validate the implanted artificial pancreas in swine studies, and to perform a pilot study in humans.